it’s not working since hours ago, the team is stuck atm, can we have some kind of queue instad of directly cutting the connection? I know this would be like paying to have slow request again, but at the moment I really prefer that to just not being able to do anything
8 Likes
the same issue
3 Likes
Yes, this is really distressing. The key point is that it wastes precious requests.
2 Likes
Yep, we have reported the same issue here and to cursor support team but they have yet to reply to us (3 days!)
2 Likes
the same issue
why do I have a feeling when I use sonnet 4 the results are more like sonnet 3.7, at the beginning it was super good and the difference was very clear
but now I have a feeling in the background it is using sonnet 3.7 How can I confirm that I am using sonnet 4
(Request ID: 180ec5f3-7ef3-4e7e-9d26-55d96364c71e)
I get this constantly and I was able to use Sonnet 4 maybe 2-3 times in a week. I think this issue will not be resolved.
2 Likes
same error
Claude 4 is not currently enabled in the slow pool due to high demand. Please select another model, or enable usage-based pricing to get more fast requests.
1 Like
I still have the rights but my question is not about the slow request pool. I mean I want to use my current rights to use Sonnet 4 but it keeps telling me to try again and sometimes it works if I press it 50 times.
Now, multiple errors from yesterday: We’ve hit a rate limit with Vertex.
Can you please limit free users, not the paying ones?
I understand that you have added many free users and usage spikes on billlin cycle refresh, but you are hurting your loyal customer base. Are you planning to resolve it? Or should I switch to another tool?
If this does not resolve in the next few hours, I’ll be forced to switch.
3 Likes
Same boat here, was nearing the end of a very large project I am trying to accomplish using Sonnet 4 and now I have not been able to connect for almost 24 hours in normal or max modes. Request ID: c86084af-f76e-47f7-990c-3ee3539ea926
Same issue. If I upgrade to Business for $40/mo, will I still have the issue?
Same here, and I’m a paid user. Honestly, with the charges being incurred for premium messages and the fact that as a paid user they can just ‘turn off’ models that simply exist at Anthropic and Anthropic doesn’t turn off themself, this is such a scummy move that I’m considering moving to Claude Code.
there is no difference between 20$ & 40$ plan, IMO 
1 Like
And now Gemini 2.5 is also exceeding API limits so we don’t have Claude or Gemini. OpenAI models are all that are working for me at the moment. And it doesn’t matter if I try to use my Gemini API key.
1 Like
I believe that this is related to an ongoing GCP outage.
1 Like
the project ive been working on for a week with claude 4 is now stalled waiting on a fix for this outage. super frustrating. need to get to work.
1 Like
yup, it was until today that I realized how many of my services are actually running on GCP in the background 
hm so wait until GCP is fixed?
It should be fixed now, though there are intermittent errors. If you’re easily frustrated then I would recommend just waiting a few hours for things to be truly stable again.
1 Like